|Publication number||US7913123 B2|
|Application number||US 12/210,005|
|Publication date||Mar 22, 2011|
|Priority date||Feb 11, 2005|
|Also published as||US7437617, US20060184834, US20090006825|
|Publication number||12210005, 210005, US 7913123 B2, US 7913123B2, US-B2-7913123, US7913123 B2, US7913123B2|
|Inventors||Ra'ed Mohammad Al-Omari, Alexander Erik Mericas, William John Starke|
|Original Assignee||International Business Machines Corporation|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (52), Non-Patent Citations (8), Referenced by (4), Classifications (10), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
The subject matter of the present application is related to copending U.S. application Ser. No. 11/055,281, titled “Method in a Processor for Performing In-Memory Tracing Using Existing Communication Paths”, Ser. No. 11/055,977, titled “Method in a Processor for Dynamically During Runtime Allocating Memory for In-Memory Hardware Tracing”, Ser. No. 11/055,870, titled “Synchronizing Triggering of Multiple Hardware Trace Facilities Using an Existing System Bus”, and Ser. No. 11/056,000, titled “Method, Apparatus, and Computer Program Product in a Processor for Balancing Hardware Trace Collection Among Different Hardware Trace Facilities”, all filed on even date herewith, all assigned to the assignee thereof, and all incorporated herein by reference.
This application is a continuation of application Ser. No. 11/055,845, filed Feb. 11, 2005, status allowed.
1. Technical Field
The present invention is directed to data processing systems. More specifically, the present invention is directed to a method, apparatus, and computer program product in a processor for concurrently sharing system memory among a tracing process and non-tracing processes using a programmable variable number of shared memory write buffers.
2. Description of Related Art
Making tradeoffs in the design of commercial server systems has never been simple. For large commercial systems, it may take years to grow the initial system architecture draft into the system that is ultimately shipped to the customer. During the design process, hardware technology improves, software technology evolves, and customer workloads mutate. Decisions need to be constantly evaluated and reevaluated. Solid decisions need solid base data. Servers in general and commercial servers in particular place a large demand on system and operator resources, so the opportunities to collect characterization data from them are limited.
Much of performance analysis is based on hardware-collected traces. Typically, traces provide data used to simulate system performance, to make hardware design tradeoffs, to tune software, and to characterize workloads. Hardware traces are almost operating system, application, and workload independent. This attribute makes these traces especially well suited for characterizing the On-Demand and Virtual-Server-Hosting environments now supported on the new servers.
A symmetric multiprocessing (SMP) data processing server has multiple processors with multiple cores that are symmetric such that each processor has the same processing speed and latency. An SMP system could have multiple operating systems running on different processors, which are a logically partitioned system, or multiple operating systems running on the same processors one at a time, which is a virtual server hosting environment. Operating systems divide the work into tasks that are distributed evenly among the various cores by dispatching one or more software threads of work to each processor at a time.
A single-thread (ST) data processing system includes multiple cores that can execute only one thread at a time.
A simultaneous multi-threading (SMT) data processing system includes multiple cores that can each concurrently execute more than one thread at a time per processor. An SMT system has the ability to favor one thread over another when both threads are running on the same processor.
As computer systems migrate towards the use of sophisticated multi-stage pipelines and large SMP with SMT based systems, the ability to debug, analyze, and verify the actual hardware becomes increasingly more difficult, during development, test, and during normal operations. A hardware trace facility may be used which captures various hardware signatures within a processor as trace data for analysis. This trace data may be collected from events occurring on processor cores, busses (also called the fabric), caches, or other processing units included within the processor. The purpose of the hardware trace facility is to collect hardware traces from a trace source within the processor and then store the traces in a predefined memory location.
As used herein, the term “processor” means a central processing unit (CPU) on a single chip, e.g. a chip formed using a single piece of silicon. A processor includes one or more processor cores and other processing units such as a memory controller, cache controller, and the system memory that is coupled to the memory controller.
This captured trace data may be recorded in the hardware trace facility and/or within another memory. The term “in-memory tracing” means storing the trace data in part of the system memory that is included in the processor that is being traced.
Prior art approaches to in-memory tracing used a specialized data path between the trace facility and the memory controller. For example,
A multiplexer 1508 selects either the signal from fabric bus controller/bus 1506 or the signal from trace facility 1510. When in a normal, non-tracing, processing mode, multiplexer 1508 selects the signal from fabric bus controller/bus 1506. When in a trace mode when trace facility is collecting and needs to store traces in system memory 1502, multiplexer 1508 selects the signal from trace facility 1510. Thus, as is clear from
There are problems with the prior art method. When in a trace mode, memory controller 1501 is dedicated to trace facility 1510. While memory controller 1501 is dedicated to trace facility 1510, it is precluded from being used for any other purpose. This is a significant limitation, particularly in systems that have only one memory controller. In systems with only one memory controller, the system must be dedicated to the trace function and cannot perform any other work that would require the use of system memory 1502 when in trace mode.
In addition, the prior art system requires that in-memory tracing be completed using the system memory 1502 that is part of the processor 1500 that is being traced. The trace data captured by trace facility 1510 cannot be stored in any memory other than system memory 1502.
In addition to the limitations described above, the prior art requires that the system be booted to a trace mode instead to a normal mode when tracing is desired. In the prior art systems, the memory had to be allocated to store traces prior to the initial program load (IPL) being completed.
Referring again to block 1604, if a determination is made that trace data is to be stored, the process passes to block 1612 which depicts allocating memory for storing traces. The dedicated memory will be a fixed size throughout the trace process. The size of the dedicated memory will not be able to be changed without rebooting the system and executing another IPL process.
A memory controller is dedicated to the trace process as described above. Because the memory controller is dedicated to the trace process, the rest of the processor, other than the trace facility, loses the ability to write to the memory that is controller by the dedicated memory controller.
Thereafter, block 1614 illustrates executing the IPL process to trace. This is a different IPL process than the normal IPL process executed as depicted by block 1606. For example, during the trace IPL process, multiplexers are set for tracing. Next, block 1616 depicts capturing traces. Thereafter, block 1618 illustrates a determination of whether or not tracing is finished. If a determination is made that tracing is not finished, the process passes back to block 1616. Referring again to block 1618, if a determination is made that tracing has finished, the process passes to block 1620 which illustrates a determination of whether or not to start normal processing. If a determination is made not to start normal processing, the process passes back to block 1620. If a determination is made to start normal processing, the process passes back to block 1602.
Therefore, a need exists for a method, apparatus, and computer program product in a processor for concurrently sharing system memory among a tracing process and non-tracing processes using a programmable variable number of shared memory write buffers.
An apparatus and computer program product are disclosed for, in a processor, concurrently sharing a memory controller among a tracing process and non-tracing processes using a programmable variable number of shared memory write buffers. A hardware trace facility captures hardware trace data in a processor. The hardware trace facility is included within the processor. The hardware trace data is transmitted to a system memory utilizing a system bus. The system memory is included within the system. The system bus is capable of being utilized by processing units included in the processing node while the hardware trace data is being transmitted to the system bus. Part of system memory is utilized to store the trace data. The system memory is capable of being accessed by processing units in the processing node other than the hardware trace facility while part of the system memory is being utilized to store the trace data.
The above as well as additional objectives, features, and advantages of the present invention will become apparent in the following detailed written description.
The novel features believed characteristic of the invention are set forth in the appended claims. The invention itself, however, as well as a preferred mode of use, further objectives and advantages thereof, will best be understood by reference to the following detailed description of an illustrative embodiment when read in conjunction with the accompanying drawings, wherein:
A preferred embodiment of the present invention and its advantages are better understood by referring to the figures, like numerals being used for like and corresponding parts of the accompanying figures.
The present invention is a method, apparatus, and computer program product in a processor for concurrently sharing system memory among a tracing process and non-tracing processes using a programmable variable number of shared memory write buffers.
The HTM looks like any other processing unit in the processor to the fabric. It uses the same data and addressing scheme, protocols, and coherency used by the other processing units in the processor. Therefore, there is no need for extra wiring or side band signals. There is no need for a special environment for verification since it will be verified with the standard existing verification functions.
The HTM captures hardware trace data in the processor and transmits it to a system memory utilizing a system bus. The system bus, referred to herein as the fabric and/or fabric bus controller and bus, is capable of being utilized by processing units included in the processor while the hardware trace data is being transmitted to the system bus. A standard bus protocol is used by these processing units to communication with each other via the standard existing system bus.
According to the present invention, the hardware trace facility, i.e. the HTM, is coupled directly to the system bus. The memory controllers are also coupled directly to the system bus. The HTM uses this standard existing system bus to communicate with a particular memory controller in order to cause the memory controller to store hardware trace data in the system memory that is controlled by that memory controller.
The HTM transmits its hardware trace data using the system bus. The hardware trace data is formatted according to the standard bus protocol used by the system bus and the other processing units. The hardware trace data is then put out on the bus in the same manner and format used for to transmit all other information.
The memory controller(s) snoop the bus according to prior art methods.
According to the present invention, when trace data is destined for a particular memory controller, the trace data is put on the bus as bus traffic that is formatted according to the standard bus protocol. The particular memory controller is identified in this bus traffic. The memory controller will then retrieve the trace data from the bus and cause the trace data to be stored in the memory controlled by this memory controller.
In a preferred embodiment, a data processing system includes multiple nodes. Each node includes four separate processors. Each processor includes two processing cores and multiple processing units that are coupled together using a system bus. The system busses of each processor in each node are coupled together. In this manner, the processors in the various nodes can communicate with processors in other nodes via their system busses following the standard bus protocol.
One or more memory controllers are included in each processor. The memory controller that is identified by the bus traffic can be any memory controller in the system. Each memory controller controls a particular system memory. Because the standard system bus and bus protocol are used by the HTM, the trace data can be stored in the system memory in the processor which includes the HTM that captured trace data by identifying, in the bus traffic, the memory controller that controls this memory. The trace data can instead be stored in a system memory in another processor in this node or in any other node by identifying, in the bus traffic, a memory controller in another processor in this node or a memory controller in a different node.
Also according to the present invention, a memory controller does not need to be dedicated to the HTM. A particular memory controller can be used by the HTM for storing trace data to the system memory that is controlled by the memory controller at the same time that same memory controller is being used by another process to store data to the system memory for that process.
Each memory controller controls a particular number of write buffers. The HTM can request a variable number of write buffers to be allocated to the HTM. This variable number is determined by a control routine that first identifies the trace that is to be collected. Some traces generate more trace data than other traces. For example, a fabric trace generates a lot of trace data and will need more write buffers than another kind of trace.
The control routine sets the number of write buffers to be requested based on the type of trace. Once the type of trace is determined, the control routine determines the typical bandwidth used by that type of tracer the typical number of records generated by that type of tracer and the size of memory needed to store the typical number of records. The control routine then sets the number of write buffers to a particular number that will both optimize the transfer of trace data while optimizing the performance of the processor with respect to the other processing that the processor continues to do.
The control routine sets the size of system memory to be requested by storing the size in one of the registers in the HTM. The HTM will then request that size of memory to be allocated to the HTM. The size of requested memory to be allocated is variable and can change each time the HTM is enabled to capture traces.
Prior to starting a trace, the HTM will be configured to capture a particular trace. The HTM will first request that system memory be allocated to the HTM for storing the trace data it is about to collect. This memory is then allocated to the HTM for its exclusive use. The memory may be located in any system memory in the data processing system regardless of in which processor the trace data is originating.
According to the present invention, the memory controller is connected directly to the fabric bus controller. The memory controller is not coupled to the fabric bus controller through a multiplexer.
The trace facility, i.e. the hardware trace macro (HTM), is coupled directly to the fabric bus controller as if it were any other type of storage unit, e.g. an L3 cache controller, an L2 cache controller, or a non-cacheable unit. The HTM uses cast out requests to communicate with the memory controllers. A cast out request is a standard type of request that is used by the other processing units of the processor to store data in the memory. Processing units in one processor can cast out data to the system memory in that processor on to memory in other processors in this node or other processors in other nodes.
These cast out requests consist of two phases, address and data requests. These cast out requests are sent to the fabric bus controller which places them on the bus. All of the processing units that are coupled directly to the bus snoop the bus for address requests that should be processed by that processing unit. Thus, the processing units analyze each address request to determine if that processing unit is to process the request. For example, an address request may be a request for the allocation of a write buffer to write to a particular memory location. In this example, each memory controller will snoop the request and determine if it controls the system memory that includes the particular memory location. The memory controller that controls the system memory that includes the particular memory location will then get the cast out request and process it.
A cast out data request is used by the HTM to notify the fabric bus controller that the HTM trace buffer has trace data to be copied. The fabric bus controller then needs to copy the data. The fabric bus controller will use a tag, from the Dtag buffer, that includes an identification of a particular memory controller and a write buffer. The fabric bus controller then copies the data to the specific memory controller write buffer, which is identify by the tag.
Because the HTM uses cast out requests to communicate with the memory controllers, any memory controller, and thus any system memory, can be used for storing trace data. The fabric bus controller/bus transmits requests to the processing units in the processor that controls the HTM and also transmits requests to other processors in the same node as this processor and to other nodes as well. Therefore, a system memory in this processor, in another processor in this node, or in a processor in another node, can be used for storing trace data from this HTM.
Each processor, such as processor 10, includes two cores, e.g. cores 12, 14. A node is a group of four processors. For example, processor 10, processor 52, processor 54, and processor 56 are all part of node 58. There are typically multiple nodes in a data processing system. For example, node 58, node 60, and node 62 are all included in data processing system 64. Thus, communications links 38 are used to communicate among processors 10, 52, 54, and 56. Communications links 40 are used to communicate among processors in nodes 58, 60, and 62.
Although connections are not depicted in
Processor core 100 includes level one (L1) instruction and data caches (I Cache and D Cache) 102 and 104, respectively, each having an associated memory management unit (I MMU and D MMU) 106 and 108. As shown in
The dispatch unit within instruction unit 116 dispatches instructions as appropriate to execution units such as system unit 118, integer unit 120, floating point unit 122, or load/store unit 124. System unit 118 executes condition register logical, special register transfer, and other system instructions. Integer or fixed-point unit 120 performs add, subtract, multiply, divide, shift or rotate operations on integers, retrieving operands from and storing results in integer or general purpose registers (GPR File) 126. Floating point unit 122 performs single precision and/or double precision multiply/add operations, retrieving operands from and storing results in floating point registers (FPR File) 128. VMX unit 134 performs byte reordering, packing, unpacking, and shifting, vector add, multiply, average, and compare, and other operations commonly required for multimedia applications.
Load/store unit 124 loads instruction operands from data caches 104 into integer registers 126, floating point registers 128, or VMX unit 134 as needed, and stores instructions results when available from integer registers 126, floating point registers 128, or VMX unit 134 into data cache 104. Load and store queues 130 are utilized for these transfers from data cache 104 to and from integer registers 126, floating point registers 128, or VMX unit 134. Completion unit 132, which includes reorder buffers, operates in conjunction with instruction unit 116 to support out-of-order instruction processing, and also operates in connection with rename buffers within integer and floating point registers 126 and 128 to avoid conflict for a specific register for instruction results. Common on-chip processor (COP) and joint test action group (JTAG) unit 136 provides a serial interface to the system for performing boundary scan interconnect tests.
The architecture depicted in
Snoop stage 300 is used for collecting raw traces from different sources and then formatting the traces into multiple 128-bit frames. Each frame has a record valid bit and double record valid bit. The double record valid bit is used to identify if both the upper halves, e.g. bits 0-63, and the lower halves, e.g. bits 64-127, of the trace record are valid. If both bits, valid and double valid bits, are set to “1”, both halves are valid. If the double valid bit is set to “0”, only the upper half, i.e. bits 0-63, is valid. If both are set to “0” then none of the halves has valid data.
Snoop stage 300 snoops the traffic on fabric 36. Snoop stage 300 retrieves trace data from fabric 36 according to the filter and mode settings in HTM 34.
The trace data inputs to snoop stage 300 are the five hardware trace sources 310, select trace mode bits, capture mode bit, and filter mode bits 312. The outputs from this stage are connected to cast out stage 302. The outputs are a 128 bit trace record 314, a valid bit 316, and a double record valid bit 318.
There are five hardware trace sources: a core trace, a fabric trace, i.e. FBC trace, an LLATT trace, a PMU trace, and a thermal trace.
The core trace is an instruction trace for code streams that are running on a particular core.
The FBC trace is a fabric trace and includes all valid events, e.g. requests and responses, that occur on the fabric bus.
The LLATT trace is a trace from an L2 cache that is included within a processor. The LLATT trace includes load and store misses of the L1 cache generated by instruction streams running on a particular core.
The PMU trace is a performance monitor trace. It includes traces of events from the L3 cache, each memory controller, the fabric bus controller, and I/O controller.
The thermal trace includes thermal monitor debug bus data.
Trace cast out stage 302 is used for storing the trace record received from snoop stage 300 to one of the system memories 410, 420 or to another system memory in another processor that is either in this or another node. Trace cast out stage 302 is also responsible for inserting the proper stamps 320 into the trace data and managing trace buffer 306. Trace cast out stage 302 includes interfaces to fabric bus controller/bus 36, snoop stage 300, trace buffer 306, Dtag buffer 308, trace triggers, operation modes and memory allocation bits, and status bits.
Multiple different types of stamps are generated by stamps 320. A start stamp is created in the trace buffer whenever there is a transition from a paused state to a tracing state. This transition is detected using the start trace trigger.
When the HTM is enabled and in the run state, a mark stamp will be inserted into the trace data when a mark trigger occurs.
A freeze stamp is created and inserted into the trace data whenever the HTM receives a freeze trace trigger.
Time stamps are generated and inserted in the trace data when certain conditions occur. For example, when valid data appears after one or more idle cycles, a time stamp is created and inserted in the trace data.
SCOM stage 304 has an SCOM satellite 304 c and SCOM registers 304 a. SCOM satellite 304 c is used for addressing the particular SCOM register. SCOM registers 304 c include an HTM collection modes register, a trace memory configuration mode register, an HTM status register, and an HTM freeze address register. SCOM registers also includes mode bits 304 b in which the various filter and capture modes are set.
Cast out stage 302 receives instructions for starting/stopping from processor cores 12, 14, SCOM stage 304, or global triggers through the fabric 36. SCOM stage 304 receives instructions that describe all of the information that is needed in order to perform a trace. This information includes an identification of which trace to receive, a memory address, a memory size, the number of write buffers that need to be requested, and a trace mode. This information is stored in registers 304 a and mode bits 304 b. This information is then provided to snoop stage 300 in order to set snoop stage 300 to collect the appropriate trace data from fabric 36.
SCOM stage 304 generates a trace enable signal 322 and signals 324.
Trace triggers 326 include a start trigger, stop trigger, pause trigger, reset trigger, freeze trigger, and an insert mark trigger. The start trigger is used for starting a trace. The stop trigger is used for stopping a trace. The pause trigger is used to pause trace collection. The reset trigger is used to reset the frozen state and reset to the top of trace buffer 306. The freeze trigger is used to freeze trace collection. The HTM will ignore all subsequent start or stop triggers while it is in a freeze state. The freeze trigger causes a freeze stamp to be inserted into the trace data. The insert mark trigger is used to insert a mark stamp into the trace data.
Trace triggers 326 may originate from the trigger unit 325. Trigger unit 325 receives trigger signals from fabric 36, one of the cores 12, 14, or SCOM stage 304.
Signals 324 include a memory allocation done (mem_alloc_done) signal, trace modes signal, memory address signal, memory size signal, and a signal “N” which is the number of pre-requested write buffers.
According to the present invention, a configurable sequential address range, controlled by one or more of the memory controllers, is configured to be allocated to the trace function. This range can be statically assigned during the initial program load (IPL) or dynamically using software. Software will support allocation and relocation of physical memory on a system that has booted and is executing.
The process of allocation and relocation includes having the firmware declare a particular memory region as “defective” and then copying the current contents of the region to a new location. The contents of the region continue to be available to the system from this new location. This particular memory region is now effectively removed from the system memory and will not be used by other processes executing on the system. This particular memory region is now available to be allocated to the hardware trace macro for its exclusive use for storing hardware trace data.
To define this memory, the software that controls the HTM will write to an SCOM register using calls to the hypervisor. This SCOM register has a field that is used to define the base address and the size of the requested memory. The HTM will then wait until a Mem_Alloc_Done signal is received before it starts using the memory.
After enabling the HTM and allocating system memory in which to store trace data, the HTM will start the process of collecting trace data by selecting one of its inputs, i.e. inputs 310, to be captured. The trace routine that is controlling the HTM will define the memory beginning address, the memory size, and the maximum number of write buffers that the HTM is allowed to request before it has trace data to store.
To initiate the write buffer allocation process, the HTM will serially drive a series of cast out requests to the fabric controller bus, one for each number of write buffers that are allowed. If no write buffers are pre-allocated, the HTM will send a cast out request each time it has accumulated a cache line of data. A cache line of data is preferably 128 bytes of data.
The HTM will keep a count of the number of write buffers currently allocated to the HTM. Upon receiving a response from the fabric bus controller that a write buffer has been allocated to the HTM, the HTM will increment the count of the number of allocated buffers. This response will include routing information that identifies the particular memory controller that allocated the write buffer and the particular write buffer allocated. The HTM will save the routing information received from the fabric bus controller as a tag in Dtag buffer 308. This information will be used when the HTM generates a cast out data request that indicates that the HTM has trace data in trace buffer 306 that is ready to be stored in the system memory. If the response from the fabric bus controller indicates that a write buffer was not allocated, the HTM will retry its request.
When the HTM receives a start trace trigger, the HTM will begin collecting the trace that is selected using signals 312. Multiplexer 311 is controlled by signals 312 to select the desired trace. The trace data is then received in trace record 314 and then forwarded to trace buffer 306. At the start of the trace, prior to saving any trace data, a start stamp from stamps 320 is saved in trace buffer 306 to indicate the start of a trace.
When the HTM has collected 128 bytes of data, including trace data and any stamps that are stored, the HTM will send a cast out data request signal to the fabric bus controller if there is at least one write buffer allocated to the HTM. Otherwise, the HTM will request the allocation of a write buffer, wait for that allocation, and then send the cast out data request. Trace buffer 306 is capable of holding up to four cache lines of 128 bytes each. Once trace buffer 306 is full, it will start dropping these trace records. An 8-bit counter increments for every dropped record during this period of time that the buffer is full. If the 8-bit counter overflows, a bit is set and the counter rolls over and continues to count. When the buffer frees up, a timestamp entry is written before the next valid entry is written.
The fabric bus controller will then copy the data out of trace buffer 306 and store it in the designated write buffer. The HTM will then decrement the number of allocated write buffers.
When the HTM receives a stop trace trigger, the HTM will stop tracing.
Any number of write buffers may be used although only four are depicted. For example, in the preferred embodiment there are 12 write buffers for each memory controller.
As is depicted by
A variable number of write buffers may be allocated to a particular process. In addition, some of these write buffers may be allocated to one process while the remaining write buffers are allocated to another process. For example, three write buffers can be allocated to the HTM for storing trace data while the remaining nine write buffers are allocated to another process. The memory controller that controls these write buffers would then be responding to memory requests concurrently from the HTM and from the other process. Thus, neither the memory controller nor its write buffers are dedicated to the HTM for tracing.
Thereafter, block 508 illustrates the trace control routine sending a notice to the hypervisor to control tracing, such as by stopping, pausing, or performing another trace function as desired by the trace control program. The process then terminates as depicted by block 510.
The process then passes to block 606 which illustrates the hypervisor storing the specified size of memory to be requested in a register in the SCOM in the HTM. Next, block 608 depicts the hypervisor storing other information in the SCOM registers that is needed to control tracing. The process then terminates as depicted by block 610.
Block 708, then, depicts the trace enable signal in the HTM being turned on. Next, block 710 illustrates the HTM requesting an allocation of a specific size of system memory to be used to store trace data. The HTM requests the allocation of the size of memory that is identified within registers 304 a. The system memory can be anywhere in the system. The system memory can be on the same processor as the HTM, on a different processor, or in a node that is different from the node that includes the HTM.
Block 712, then, illustrates the HTM receiving a base address of the memory allocated to the HTM by the hypervisor after the hypervisor has allocated the memory. Next, block 714 depicts setting a number of write buffers to be maintained for the HTM. This is the number of write buffers that should always be allocated to the HTM. This number is stored in registers 304 a. The process then passes to block 716 which illustrates setting the current address equal to the base address. Thereafter, block 718 depicts maintaining a count of the number of write buffers that are currently allocated to the HTM.
Thereafter, block 720 illustrates a determination of whether or not the HTM has received a stop trigger. If a determination is made that the HTM has received a stop trigger, the process terminates as depicted by block 722. Referring again to block 720, if a determination is made that the HTM has not received a stop trigger, the process passes to block 724 which illustrates a determination of whether or not there is a need to request additional write buffers to be allocated to the HTM. If the count of write buffers that are currently allocated to the HTM is not below the set number of write buffers to be maintained, the process passes back to block 724. In this case, the number of write buffers currently allocated to the HTM is equal to the number of write buffers (set in block 714) to be maintained.
Referring again to block 724, if the count of write buffers that are currently allocated to the HTM is below the set number of write buffers to be maintained, the process passes to block 726 which depicts the HTM sending a cast out request requesting the allocation of a write buffer to use to write to the current address in the allocated memory.
Block 728, then, illustrates a determination of whether or not the HTM has received a retry signal. If a determination is made that the HTM has received a retry signal, the process passes back to block 726. If a determination is made that the HTM has not received a retry signal, the process passes to block 730 which depicts the HTM receiving a response from the memory controller that identifies a particular write buffer. This is the write buffer that was allocated to the HTM.
Next, block 732 illustrates saving the write buffer number and a memory controller identifier, such as the memory controller number, in the Dtag buffer as a tag. Thereafter, block 734 depicts adding one to the count of the number of write buffers that are currently allocated to the HTM. Block 736, then, illustrates setting the current address equal to the current address plus 128 bytes. The process then passes back to block 720.
The process then passes to block 806 which illustrates the hypervisor marking the selected memory as “defective”. Next, block 808 depicts the hypervisor copying the contents of the selected memory to another memory location. Block 810, then, illustrates all accesses to the selected memory now being routed to the new memory location. Thereafter, block 812 depicts the hypervisor returning a notice, such as a “mem_alloc_done” signal, to the HTM that the memory has been allocated. This notice also includes the base address of the allocated memory. A bit is then set in mode bits 304 to indicate that the memory has been allocated. In addition, the base address is stored in registers 304 a. The process then terminates as illustrated by block 814.
Block 908, then, depicts determining the size of memory that will be needed to store the typical number of records generated by this type of trace. The process then passes to block 910 which illustrates determining the number of write buffers that will be needed to accommodate the typical trace bandwidth while optimizing the remaining processing of the system. Block 912, then, depicts setting the number of write buffers equal to the number determined as being needed to accommodate this trace bandwidth. This number is variable and can range anywhere between one write buffer to the total number of write buffers that are supported by one memory controller. In the preferred embodiment, twelve write buffers are supported by each memory controller. The process then terminates as illustrated by block 914.
Thereafter, block 1008 depicts a determination of whether or not the buffer is full. If a determination is made that the buffer is full, the process passes to block 1010 which illustrates dropping traces and incrementing the drop counter. The process then passes back to block 1008. Referring again to block 1008, if a determination is made that the buffer is not full, the process passes to block 1012 which depicts a determination of whether or not a stop trigger has been received. If a determination is made that a stop trigger has been received, the process terminates as illustrated by block 1014. Referring again to block 1012, if a determination is made that a stop trigger has not been received, the process passes to block 1016 which depicts collecting trace data.
The process then passes to block 1018 which illustrates a determination of whether or not a cache line, e.g. 128 bytes, has been collected and a write buffer allocated. If a determination is made that either 128 bytes has not been collected yet or a write buffer has not been allocated, the process passes back to block 1016. If a determination is made that 128 bytes has been collected and a write buffer has been allocated, the process passes to block 1020 which depicts sending a cast out data request to the fabric bus controller. The cast out data request includes an address in the trace buffer where the trace data is located. The cast out data request also includes Dtag information, such as a tag, for the next write buffer. The tag includes an identifier of a memory controller, such as a memory controller number, and an identifier of a write buffer, such as a write buffer number.
Block 1022, then, depicts the data being copied out of the trace buffer. Next, block 1024 illustrates the HTM receiving an acknowledgement from the fabric bus controller that the data has been copied to the write buffer that was identified in the cast out data request. Thereafter, block 1026 depicts subtracting one from the count of the number of write buffers that are currently allocated to the HTM. The process then passes back to block 1008.
The process then passes to block 1106 which illustrates a determination of whether or not one of this memory controller's write buffers is currently available to be allocated. If a determination is made that none of this memory controller's write buffers are currently available to be allocated, the process passes to block 1108 which depicts the memory controller sending a retry signal to the HTM. The process then passes back to block 1102.
Referring again to block 1106, if a determination is made that one of this memory controller's write buffers is currently available to be allocated, the process passes to block 1110 which depicts the memory controller allocating a write buffer. This write buffer is allocated for use for the specified memory address that was included in the cast out request. This write buffer is associated with this particular memory address as long as the write buffer is allocated for this memory address. Next, block 1112 illustrates the memory controller sending a response to the HTM that acknowledges the allocation of a write buffer and that identifies the particular write buffer that was allocated to the HTM. This acknowledgement also identifies this particular memory controller. For example, the acknowledgement can include the memory controller number and the write buffer number. The process then passes back to block 1102.
The cast out data request includes the address in the trace buffer where the data is located. The cast out data request also includes routing information in the form of a tag from the Dtag buffer. The tag from the Dtag buffer includes a memory controller identifier, such as a memory controller number, and a write buffer identifier, such as a write buffer number. Thus, the tag from the Dtag buffer can be used to locate a particular memory controller and then a particular write buffer.
The process then passes to block 1206 which depicts the fabric bus controller copying the data from the trace buffer that is included in the HTM to the write buffer that was identified in the cast out data request. Next, block 1208 illustrates the fabric bus controller sending an acknowledgement to the HTM that the data has been copied. The process then passes back to block 1202.
The process then passes to block 1306 which illustrates the fabric bus sending the request out over the bus. Thereafter, block 1308 depicts the fabric bus controller sending an acknowledgement to the HTM that the request has been sent. Block 1310, then, illustrates the fabric bus controller getting a response from the memory controller and forwarding that response to the HTM. The process then passes back to block 1302.
Data processing system 1420 includes logically partitioned platform 1450. Platform 1450 includes partitioned hardware 1452, partition management firmware, also called a hypervisor 1454, and partitions 1456-1459. Operating systems 1461-1464 exist within partitions 1456-1459. Operating systems 1461-1464 may be multiple copies of a single operating system or multiple heterogeneous operating systems simultaneously run on platform 1450.
Partitioned hardware 1452 includes a plurality of SMT-capable processors 1465-1468, a plurality of system memory units 1470-1473, a plurality of input/output (I/O) adapters 1474-1481, and a storage unit 1482. Each of the processors 1465-1468, memory units 1470-1473, NVRAM storage 1483, and I/O adapters 1474-1481 may be assigned to one of multiple partitions 1456-1459. Partitioned hardware 1452 also includes service processor 1490. A non-volatile memory device 1491, such as an NVRAM device, is included within service processor 1490.
Partition management firmware (hypervisor) 1454 performs a number of functions and services for partitions 1456-1459 to create and enforce the partitioning of logically partitioned platform 1450. Hypervisor 1454 is a firmware implemented virtual machine identical to the underlying hardware. Firmware is “software” stored in a memory chip that holds its content without electrical power, such as, for example, read-only memory (ROM), programmable ROM (FROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), and non-volatile random access memory (non-volatile RAM). Thus, hypervisor 1454 allows the simultaneous execution of independent OS images 1461-1464 by virtualizing all the hardware resources of logically partitioned platform 1450. Hypervisor 1454 may attach I/O devices through I/O adapters 1474-1481 to single virtual machines in an exclusive mode for use by one of OS images 1461-1464.
A hardware management console (HMC) 1494 may be coupled to service processor 1490 in data processing system 1420. HMC 1494 is a separate computer system that is coupled to service processor 1490 and may be used by a user to control various functions of system 1420 through service processor 1490.
It is important to note that while the present invention has been described in the context of a fully functioning data processing system. Those of ordinary skill in the art will appreciate that the processes of the present invention are capable of being distributed in the form of a computer readable medium of instructions and a variety of forms and that the present invention applies equally regardless of the particular type of signal bearing media actually used to carry out the distribution. Examples of computer readable media include recordable-type media, such as a floppy disk, a hard disk drive, a RAM, CD-ROMs, DVD-ROMs, and transmission-type media, such as digital and analog communications links, wired or wireless communications links using transmission forms, such as, for example, radio frequency and light wave transmissions. The computer readable media may take the form of coded formats that are decoded for actual use in a particular data processing system.
The description of the present invention has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the invention in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiment was chosen and described in order to best explain the principles of the invention, the practical application, and to enable others of ordinary skill in the art to understand the invention for various embodiments with various modifications as are suited to the particular use contemplated.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4205370||Apr 16, 1975||May 27, 1980||Honeywell Information Systems Inc.||Trace method and apparatus for use in a data processing system|
|US4399506 *||Oct 6, 1980||Aug 16, 1983||International Business Machines Corporation||Store-in-cache processor means for clearing main storage|
|US4590550||Jun 29, 1983||May 20, 1986||International Business Machines Corporation||Internally distributed monitoring system|
|US4835675||Oct 19, 1987||May 30, 1989||Mitsubishi Denki Kabushiki Kaisha||Memory unit for data tracing|
|US5206948||Dec 22, 1989||Apr 27, 1993||Bull Hn Information Systems Inc.||Bus monitor with means for selectively capturing trigger conditions|
|US5210862||Dec 22, 1989||May 11, 1993||Bull Hn Information Systems Inc.||Bus monitor with selective capture of independently occuring events from multiple sources|
|US5642478||Dec 29, 1994||Jun 24, 1997||International Business Machines Corporation||Distributed trace data acquisition system|
|US5655103 *||Feb 13, 1995||Aug 5, 1997||International Business Machines Corporation||System and method for handling stale data in a multiprocessor system|
|US5715430||Dec 18, 1996||Feb 3, 1998||Kabushiki Kaisha Toshiba||Multiprocessor computer system and a method for memory allocation to optimize cache coherency within the system|
|US5764885||Dec 19, 1994||Jun 9, 1998||Digital Equipment Corporation||Apparatus and method for tracing data flows in high-speed computer systems|
|US5765198||Feb 1, 1996||Jun 9, 1998||Cray Research, Inc.||Transparent relocation of real memory addresses in the main memory of a data processor|
|US5862316||Jul 1, 1996||Jan 19, 1999||Sun Microsystems, Inc.||Multiprocessing system having coherency-related error logging capabilities|
|US5890217||Feb 7, 1996||Mar 30, 1999||Fujitsu Limited||Coherence apparatus for cache of multiprocessor|
|US5897658||Mar 13, 1997||Apr 27, 1999||International Business Machines Corporation||Method and apparatus for protecting portions of memory by providing access requests to a communications area for processing by a hidden server|
|US6134676||Apr 30, 1998||Oct 17, 2000||International Business Machines Corporation||Programmable hardware event monitoring method|
|US6282615||Nov 9, 1999||Aug 28, 2001||International Business Machines Corporation||Multiprocessor system bus with a data-less castout mechanism|
|US6330591||Mar 9, 1998||Dec 11, 2001||Lsi Logic Corporation||High speed serial line transceivers integrated into a cache controller to support coherent memory transactions in a loosely coupled network|
|US6401174||Sep 4, 1998||Jun 4, 2002||Sun Microsystems, Inc.||Multiprocessing computer system employing a cluster communication error reporting mechanism|
|US6421756||May 6, 1999||Jul 16, 2002||International Business Machines Corporation||Buffer assignment for bridges|
|US6457144||Dec 8, 1998||Sep 24, 2002||International Business Machines Corporation||System and method for collecting trace data in main storage|
|US6539500||Oct 28, 1999||Mar 25, 2003||International Business Machines Corporation||System and method for tracing|
|US6678838||Aug 23, 1999||Jan 13, 2004||Advanced Micro Devices, Inc.||Method to track master contribution information in a write buffer|
|US6857083||Dec 18, 2000||Feb 15, 2005||International Business Machines Corporation||Method and system for triggering a debugging unit|
|US6898736||Oct 31, 2001||May 24, 2005||International Business Machines Corporation||Dynamic sizing logic for dump list generation|
|US6912673||Feb 1, 2002||Jun 28, 2005||Cradle Technologies, Inc.||Bus analyzer unit with programmable trace buffers|
|US6920521||Oct 10, 2002||Jul 19, 2005||International Business Machines Corporation||Method and system of managing virtualized physical memory in a data processing system|
|US6985848||Mar 2, 2001||Jan 10, 2006||Texas Instruments Incorporated||Obtaining and exporting on-chip data processor trace and timing information|
|US7007205||Feb 15, 2001||Feb 28, 2006||Silicon Graphics, Inc.||Method and apparatus for recording trace data in a microprocessor based integrated circuit|
|US7010726||Mar 1, 2001||Mar 7, 2006||International Business Machines Corporation||Method and apparatus for saving data used in error analysis|
|US7080289||Oct 10, 2001||Jul 18, 2006||Arm Limited||Tracing multiple data access instructions|
|US7152186||Aug 4, 2003||Dec 19, 2006||Arm Limited||Cross-triggering of processing devices|
|US7200776||Nov 20, 2002||Apr 3, 2007||Sun Microsystems, Inc.||System and method for generating trace data in a computing system|
|US7213169||Apr 3, 2003||May 1, 2007||International Business Machines Corporation||Method and apparatus for performing imprecise bus tracing in a data processing system having a distributed memory|
|US7269756||Mar 24, 2004||Sep 11, 2007||Intel Corporation||Customizable event creation logic for hardware monitoring|
|US20020147943||Apr 10, 2001||Oct 10, 2002||Slaugh Chad H.||System and method for allocating logic analyzer hardware resources|
|US20030145255||Jan 15, 2002||Jul 31, 2003||Harty Anthony Walter||Hierarchical multi-component trace facility using multiple buffers per component|
|US20030196141||Apr 21, 2003||Oct 16, 2003||Mark Shaw||Hierarchy of fault isolation timers|
|US20040019828||Jul 25, 2002||Jan 29, 2004||Gergen Joseph P.||Method and apparatus for debugging a data processing system|
|US20040199722||Apr 3, 2003||Oct 7, 2004||International Business Machines Corp.||Method and apparatus for performing bus tracing in a data processing system having a distributed memory|
|US20040199823||Apr 3, 2003||Oct 7, 2004||International Business Machines Corporation||Method and apparatus for performing imprecise bus tracing in a data processing system having a distributed memory|
|US20050102477 *||Jul 8, 2004||May 12, 2005||Hitachi, Ltd.||Multiprocessor system|
|US20050172293||Apr 21, 2004||Aug 4, 2005||Network Appliance, Inc.||Method and apparatus for allocating resources in a shared resource processor|
|US20050268171||Apr 22, 2005||Dec 1, 2005||House Richard W||Enterprise test data management system utilizing hierarchical test data models and related methods|
|US20050273672||May 18, 2004||Dec 8, 2005||Konda Dharma R||Method and system for efficiently recording processor events in host bus adapters|
|US20060064552||Oct 8, 2002||Mar 23, 2006||Ip-First, Llc.||Cache memory and method for handling effects of external snoops colliding with in-flight operations internally to the cache|
|US20060179184||Feb 10, 2005||Aug 10, 2006||International Business Machines Corporation||Method for providing low-level hardware access to in-band and out-of-band firmware|
|US20060184833||Feb 11, 2005||Aug 17, 2006||International Business Machines Corporation||Method, apparatus, and computer program product in a processor for performing in-memory tracing using existing communication paths|
|US20060184835||Feb 11, 2005||Aug 17, 2006||International Business Machines Corporation||Method, apparatus, and computer program product for synchronizing triggering of multiple hardware trace facilities using an existing system bus|
|US20060184836||Feb 11, 2005||Aug 17, 2006||International Business Machines Corporation||Method, apparatus, and computer program product in a processor for dynamically during runtime allocating memory for in-memory hardware tracing|
|US20060184837||Feb 11, 2005||Aug 17, 2006||International Business Machines Corporation||Method, apparatus, and computer program product in a processor for balancing hardware trace collection among different hardware trace facilities|
|US20070074067||Sep 29, 2005||Mar 29, 2007||Rothman Michael A||Maintaining memory reliability|
|US20070088983||Oct 20, 2004||Apr 19, 2007||Koninklijke Philips Electronics N.V.||Integrated circuit comprising a measurement unit for measuring utlization|
|1||USPTO Notice of Allowance for U.S. Appl. No. 11/055,821 dated May 1, 2008.|
|2||USPTO Notice of Allowance for U.S. Appl. No. 11/055,870 dated May 5, 2008.|
|3||USPTO Notice of Allowance for U.S. Appl. No. 11/055,977 dated Jul. 4, 2008.|
|4||USPTO Notice of Allowance for U.S. Appl. No. 12/201,557 dated Jul. 26, 2010..|
|5||USPTO Notice of Allowance for U.S. Appl. No. 12/206,967 dated Jul. 29, 2010.|
|6||USPTO Office Action for U.S. Appl. No. 11/055,821 dated Nov. 29, 2007.|
|7||USPTO Office Action for U.S. Appl. No. 11/055,870 dated Dec. 12, 2007.|
|8||Uspto Office Action for U.S. Appl. No. 11/055,977 dated Nov. 15, 2007.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8407528 *||Mar 26, 2013||Texas Instruments Incorporated||Circuits, systems, apparatus and processes for monitoring activity in multi-processing systems|
|US8595555||Jan 13, 2011||Nov 26, 2013||Xilinx, Inc.||Debugging an integrated circuit with an embedded processor|
|US8595561 *||Oct 27, 2010||Nov 26, 2013||Xilinx, Inc.||Integrated debugging within an integrated circuit having an embedded processor|
|US20100332909 *||Aug 28, 2009||Dec 30, 2010||Texas Instruments Incorporated||Circuits, systems, apparatus and processes for monitoring activity in multi-processing systems|
|U.S. Classification||714/45, 710/56, 714/39, 710/52, 714/30|
|Cooperative Classification||G06F11/2268, G06F11/348|
|European Classification||G06F11/22L, G06F11/34T6|
|Nov 16, 2011||AS||Assignment|
Owner name: BIOMET BIOLOGICS, LLC, INDIANA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:BIOMET MANUFACTURING CORP.;REEL/FRAME:027234/0380
Effective date: 20111102
|Oct 31, 2014||REMI||Maintenance fee reminder mailed|
|Mar 22, 2015||LAPS||Lapse for failure to pay maintenance fees|
|May 12, 2015||FP||Expired due to failure to pay maintenance fee|
Effective date: 20150322